web-browser

Control Chrome via CDP to navigate, evaluate JavaScript, and capture screenshots.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/j-greig/claude-skills --skill web-browser
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: web-browser
Source: https://github.com/j-greig/claude-skills/tree/main/skills/web-browser
Command: npx skills add https://github.com/j-greig/claude-skills --skill web-browser

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires puppeteer-core, and includes scripts (resource) components.

What problem does it solve?

This Skill empowers Claude to interact with web pages just like a human, eliminating the need for manual browsing, data extraction, or form filling. It automates repetitive web tasks, allowing you to gather information or perform actions on dynamic websites effortlessly.

Core Features & Use Cases

  • Remote Browser Control: Launches and controls Google Chrome/Chromium via the Chrome DevTools Protocol (CDP).
  • Web Navigation & Interaction: Navigate to URLs, execute JavaScript, click buttons, fill forms, and interact with page elements.
  • Visual & Data Capture: Take screenshots of the current viewport and extract structured information from web pages.
  • Use Case: Automate web scraping for market research, fill out online application forms, test web application flows, or gather real-time data from dynamic websites without manual intervention.

Quick Start

Use the web-browser skill to navigate to example.com and screenshot the page.

Frequently Asked Questions about web-browser

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I automate web scraping and data extraction from websites?

Web scraping automates data collection from live websites by controlling a browser remotely via Chrome DevTools Protocol. This Skill navigates pages, executes JavaScript, captures screenshots, and extracts structured information without manual browsing.

Can I use Puppeteer to control Chrome remotely for automation tasks?

Yes. Puppeteer-core enables remote control of Chrome/Chromium instances via CDP at port 9222, letting you automate navigation, form interaction, element selection, and page evaluation from your scripts.

How do I fill out forms and interact with page elements programmatically?

This Skill provides form interaction by executing JavaScript in the page context, clicking buttons, typing into fields, and selecting elements interactively—automating repetitive web tasks like application form completion and user workflows.

What's the best way to test web application flows and capture page states?

Automate web application testing by controlling a browser remotely to navigate flows, interact with UI elements, and take screenshots of page states—validating functionality without manual intervention.

Do I need to launch Chrome manually to use browser automation?

No. This Skill handles Chrome/Chromium launch and remote debugging setup automatically, establishing a CDP connection at port 9222 and managing tabs, so you only define the actions you need.

What are the limitations of web browser automation with Chrome DevTools Protocol?

Chrome/Chromium must be installed locally and remote debugging enabled at port 9222. CDP-based automation works best for deterministic workflows; dynamic or heavily JavaScript-dependent sites may require explicit waits and error handling.